Time Series Forecasting and Error Metrics | Dataford Interview Questions

Describe a time series forecasting problem you worked on and explain the difference between RMSE and MAE.

Using actual_values and forecast_values, write a SQL query that calculates both error metrics for each forecasting model. Exclude observations where either the actual or forecast value is missing.

Output

One row per model with model_name, observations_used, mae, and rmse
Order by rmse ascending, then model_name ascending

Schema

actual_values

Column	Type	Description
actual_idPK	INT	Unique actual observation identifier
observation_date	DATE	Date of the observed value
actual_value	NUMERIC(12,2)	Observed value for the date

forecast_values

Column	Type	Description
forecast_idPK	INT	Unique forecast identifier
model_name	VARCHAR(40)	Name of the forecasting model
forecast_date	DATE	Date being forecast
forecast_value	NUMERIC(12,2)	Predicted value for the date

Your question is Time Series Forecasting and Error Metrics. Start with the requirements and the two tables in the Question tab.

Run and submit as often as you like. When you're ready, talk me through your approach or go straight to the code.

Problem